Search Results

Blow moulding is one of the most important polymer processing methods for producing complex thermoplastic automotive parts. Contrary to injection molding, little attention has focused on process control and simulation of blow molding processes. Yet, there are still several problems that affect the overall success of forming these parts. Among them are thermally induced stresses, relevant shrinkage and part warpage deformations caused by inappropriate mold design and/or processing conditions. Tolerance issues are critical in automotive applications and therefore part deformation due to solidification needs to be controlled and optimized accordingly. The accurate prediction tool of part deformation due to solidification, under different cooling conditions in automotive formed parts, is important and highly suited for part designers to help achieve an efficient production.

In this work, we outline a process for traffic light detection in the context of autonomous vehicles and driver assistance technology features. For our approach, we leverage the automatic annotations from virtually generated data of road scenes. Using the automatically generated bounding boxes around the illuminated traffic lights themselves, we trained an 8-layer deep neural network, without pre-training, for classification of traffic light signals (green, amber, red). After training on virtual data, we tested the network on real world data collected from a forward facing camera on a vehicle. Our new region proposal technique uses color space conversion and contour extraction to identify candidate regions to feed to the deep neural network classifier. Depending on time of day, we convert our RGB images in order to more accurately extract the appropriate regions of interest and filter them based on color, shape and size. These candidate regions are fed to a deep neural network.

Radiator thermal cycle test is a test method to check out the robustness of a radiator. During the test, the radiator is going through transient cycles that include high and low temperature spikes. These spikes could lead to component failure and transient temperature map is the key to predict high thermal strain and failure locations. In this investigation, an accurate and efficient way of building a numerical model to simulate the transient thermal performance of the radiator is introduced. A good correlation with physical test result is observed on temperature values at various locations.

Centrifugal pumps are widely used in different thermal fluid systems in automobile industries. Computational fluid dynamics (CFD) analysis of such a thermal fluid system depends on the accurate component modeling of the system components. This paper presents CFD analysis of a centrifugal pump with two different approaches: Transient (moving grid) and the steady state - Multiple Reference Frame (MRF) methods using a commercial CFD solver Simerics MP+®. In addition, flow and pressure drop data obtained using CFD simulations of a vehicle coolant hydraulic system was compared to results from rig test data. The Transient method incorporates the real motion of the pump blades geometry and temporal flow solutions are obtained for instantaneous positions of the blade geometry. In MRF approach, the flow governing equations for the stationary zone are solved in the absolute/inertial reference frame, whereas flow in the moving zone is solved in the relative/non-inertial reference frame.

Designing an efficient vehicle coolant system depends on meeting target coolant flow rate to different components with minimum energy consumption by coolant pump. The flow resistance across different components and hoses dictates the flow supplied to that branch which can affect the effectiveness of the coolant system. Hydraulic tests are conducted to understand the system design for component flow delivery and pressure drops and assess necessary changes to better distribute the coolant flow from the pump. The current study highlights the ability of a complete 3D Computational Fluid Dynamics (CFD) simulation to effectively mimic a hydraulic test. The coolant circuit modeled in this simulation consists of an engine water-jacket, a thermostat valve, bypass valve, a coolant pump, a radiator, and flow path to certain auxiliary components like turbo charger, rear transmission oil cooler etc.